PSYCHOLOGY
2bwDGq
2bwDGq
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
Results<br />
The key to this section is using the correct statistical test to analyse your data. This is essential and can make and<br />
break the whole of your report if incorrect tests or insufficient details are present.<br />
The important part is to figure out the following:<br />
1. Type of data - nominal, ordinal or interval.<br />
2. Type of Experimental Design/Independent or related items.<br />
3. Is it a test of difference, association or correlation?<br />
There are five common non-parametric statistical tests (which don’t follow strict criterion relating to things such as<br />
the mean and standard deviation that are used:<br />
• Mann Whitney U test.<br />
• Wilcoxon Signed Rank tests.<br />
• Chi Square.<br />
• Binomial Sign test.<br />
• Spearman’s Rho.<br />
Inferential Statistics<br />
When analysing your data, you need to justify why you have used the statistical test in question for data analysis.<br />
Furthermore, you need to include the level of significance. The most common level in psychology is 0.05. This<br />
indicates the results obtained in your study are because of the 95% chance of manipulation of your independent<br />
variable, the rest 5% is due to other factors. When your results are significant, this indicates that you as a researcher<br />
can accept your alternative hypothesis and reject your null hypothesis. You would need to explicitly indicate this.<br />
In addition, all calculations of your statistical test analysis would not appear in this section. Where do you think<br />
they should go?<br />
Activity – On the basis of your aim and hypothesis, experimental design and type of data – which statistical test<br />
would you conduct for your data?<br />
15 Copyright © OCR 2015